Cell adhesion molecule 1 is upregulated in connective tissue mast cells and potentially contributes in IgE-mediated degranulation

Abstract

Mast cells are heterogeneous tissue-resident immune cells, with connective tissue mast cells (CTMCs) and mucosal mast cells in rodents exhibiting distinct phenotypes and activation profiles. Cell adhesion molecule 1 (CADM1), an immunoglobulin superfamily adhesion molecule, has been implicated in mast cell–nerve and mast cell–stromal interactions as well as in the pathogenesis of atopic dermatitis. Here, we investigated CADM1 function in CTMCs using a monoclonal antibody against the extracellular domain of CADM1, termed 3E1. CTMCs were differentiated from bone marrow–derived mast cells (BMMCs) by fibroblast coculture, where CADM1 expression was markedly upregulated compared with BMMCs. Treatment with 3E1 downregulated CADM1 expression and suppressed β-hexosaminidase release from IgE-sensitized, antigen-stimulated CTMCs by 18%, whereas BMMCs were unaffected. In addition, 3E1 reduced FM4-64–positive granule formation in activated CTMCs by 34.2% and 27.3% at 5 and 60 min, respectively. Confocal analysis further showed that 3E1 pretreatment decreased F-actin rearrangement in activated CTMCs by 66.2% at 5 min. In a passive cutaneous anaphylaxis model, intravenous 3E1 reduced dermal mast cell degranulation by 10.4%. These findings identify CADM1 as a regulator of IgE-mediated degranulation in CTMCs and demonstrate 3E1 as a valuable tool for dissecting mast cell heterogeneity in relation to tissue localization.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-40526-2.

Introduction

Mast cells are tissue-resident immune cells found predominantly in connective tissues and at mucosal surfaces, including the dermis of the skin, peritoneal cavity, and the respiratory and gastrointestinal mucosae¹. Owing to their strategic localization at sites that interface with the external environment, mast cells serve as key initiators of both innate and adaptive immune responses. In rodents, mast cells are broadly classified into two subtypes based on their anatomical distribution and phenotypic characteristics: connective tissue mast cells (CTMCs), which are primarily located in the skin and peritoneal cavity, and mucosal mast cells (MMCs), which reside mainly in the lamina propria of the intestinal mucosa¹. CTMCs are generally larger than MMCs and exhibit distinct patterns of activation. Both CTMCs and MMCs can be activated through the aggregation of high-affinity IgE receptors (FcεRI); however, only CTMCs are additionally responsive to neuropeptides such as substance P ¹. This selectivity is attributable to the exclusive expression of the Mas-related G protein–coupled receptor b2 (Mrgprb2) in CTMCs, for which substance P serves as a functional ligand¹. These differences highlight the specialized roles of mast cell subtypes in orchestrating site-specific immune responses during inflammation.

Upon activation, mast cells rapidly undergo degranulation, a process in which preformed mediators—including histamine, proteases (e.g., tryptase and chymase), and cytokines—are released from cytoplasmic secretory granules into the extracellular environment². These mediators increase vascular permeability, promote extracellular matrix remodeling, and recruit other immune cells to the site of inflammation². In the skin, mast cells participate in the pathogenesis of various inflammatory dermatoses, including atopic dermatitis, chronic urticaria, and psoriasis³. In these diseases, mast cells contribute not only to acute symptoms such as itch and edema but also to chronic inflammation and tissue remodeling through sustained release of cytokines, chemokines, and growth factors⁴. Their dual roles as initiators and amplifiers of cutaneous inflammation make mast cells attractive therapeutic targets for inflammatory skin disorders.

Cell adhesion molecule 1 (CADM1), a member of the immunoglobulin superfamily, is expressed in central and peripheral neurons, respiratory and renal epithelial cells, spermatogenic cells, and mast cells. Since identifying CADM1 as a novel adhesion molecule in mast cells in 2003 ⁵, we have investigated its physiological and pathological roles⁶. CADM1 functions as an intercellular adhesion molecule between mast cells and the mesentery⁷, fibroblasts⁸, or neurons^9,10. It promotes degranulation of bone marrow-derived mast cells (BMMCs) in response to neuronal stimulation via homophilic binding, thereby facilitating mast cell–nerve interactions^11,12. We previously demonstrated that CADM1 expression is upregulated in dermal mast cells in a hapten-induced mouse model of atopic dermatitis, and that exogenous CADM1 expression in IC-2 mast cells enhances their adhesion to dorsal root ganglion neurites and responsiveness to neuronal stimuli¹³. These findings suggest that CADM1-mediated sensory nerve–mast cell interactions may represent one of the pathogenic pathways underlying atopic dermatitis, as such interactions are known to promote pruritus in allergic conditions¹⁴. Although we have elucidated the role of CADM1 in mast cells during chronic skin diseases such as atopic dermatitis, it remains unclear whether CADM1 is also involved in acute cutaneous responses such as dermal immediate hypersensitivity.

In a series of studies, we have generated two anti-CADM1 chicken monoclonal antibodies against the extracellular domain of CADM1, named 3E1 and 9D2 ^9,15. 9D2 blocks CADM1 homophilic binding, leading to inhibition of cell–cell interactions⁹. It also partially blocks mast cell adhesion to fibroblasts¹⁶and smooth muscle cells¹⁷, suggesting inhibition of heterophilic binding between CADM1 and nectin-3 ^10,16. Recently, we reported that the combined administration of 3E1 and 9D2 induces internalization of CADM1 in CADM1-expressing tumor cells by lipid raft-mediated endocytosis¹⁸. In another study, we found that 3E1 binds strongly to neurons, including both cell bodies and neurites, and suppresses neuronal impulse transmission¹⁹. Notably, when injected subcutaneously, 3E1 localizes to nerve fibers and mast cells in the skin in association with CADM1 expression¹⁹.

In the present study, we aimed to explore the potential involvement of CADM1 in CTMC degranulation. To this end, we examined the effects of the anti-CADM1 antibody 3E1 in cultured CTMCs and in a murine model of dermal immediate hypersensitivity. We found that 3E1 treatment reduced IgE-mediated degranulation of mast cells both in vivo and in vitro, suggesting that CADM1 may contribute to mast cell activation during acute allergic skin responses.

Results

CADM1 was upregulated in CTMCs and 3E1 suppressed IgE-mediated degranulation of CTMCs

BMMCs were generated from mouse bone marrow cell suspensions as described previously¹¹. CTMCs were obtained by coculturing BMMCs with NIH/3T3 fibroblasts in stem cell factor (SCF)-supplemented medium without IL-3, as previously reported²⁰. The purity of mast cells in both BMMC and CTMC cultures was confirmed by flow cytometric analysis of c-Kit and FcεRI expression (BMMC: 96%, CTMC: 97%; Supplementary Figure S1A). Compared with BMMCs, CTMCs exhibited intense metachromatic staining with toluidine blue (Fig. 1A) and intense staining with safranin O (Supplementary Figure S1B), consistent with previous reports^21,22. Importantly, CADM1 protein expression was upregulated in CTMCs, showing a 2.6-fold increase over levels in BMMCs (Fig. 1B).

Fig. 1.

Development of connective tissue mast cells. (A) Connective tissue mast cells (CTMCs) were generated by coculturing bone marrow-derived mast cells (BMMCs) with NIH/3T3 fibroblasts in stem cell factor-supplemented medium without IL-3. Successful differentiation of CTMCs was confirmed by intense metachromatic staining with toluidine blue. Scale bar: 100 μm (top), 50 μm (bottom). (B) CADM1 expression levels in CTMCs and BMMCs were compared by immunoblotting. **: p ≤ 0.01 vs. BMMCs.

Consistent with the results of toluidine blue and safranin O staining (see above), differentiation of BMMCs into CTMCs was associated with upregulation of chymase, whereas the protein levels of tryptase and FcεRI remained unchanged (Fig. 2A). Following a 6-hour treatment with chicken–mouse chimeric 3E1 (cm3E1), CADM1 protein levels in CTMCs were reduced to 37.6% of those observed in normal mouse IgG (mIgG)-treated cells, whereas CADM1 expression in BMMCs was unaffected (Fig. 2A). In contrast, the protein levels of other key mast cell molecules, including tryptase, chymase, and FcεRI, were unchanged by cm3E1 treatment in both CTMCs and BMMCs (Fig. 2A). We next assessed mast cell degranulation in these cells using β-hexosaminidase release assays after stimulation with either anti-2,4-dinitrophenyl (DNP) IgE/DNP antigen or substance P (Fig. 2B). In BMMCs, cm3E1 treatment did not significantly affect β-hexosaminidase release under either stimulation condition (Fig. 2B). In CTMCs, DNP antigen stimulation increased β-hexosaminidase release from approximately 6% in resting cells to 34% in mIgG-treated controls. In contrast, cm3E1-treated CTMCs exhibited a significantly attenuated response, reaching only 28%, representing a 18% reduction in degranulation compared with that of mIgG-treated cells. Substance P-induced degranulation in CTMCs was unaffected by cm3E1 treatment (Fig. 2B).

Fig. 2.

Effects of 3E1 on CADM1 expression and degranulation in bone marrow-derived mast cells and connective tissue mast cells. (A) Bone marrow-derived mast cells (BMMCs) and connective tissue mast cells (CTMCs) were treated with 1 µg/ml chicken–mouse 3E1 (cm3E1) or control mouse IgG (mIgG) for 6 h. CADM1 protein levels and bound antibody were analyzed by immunoblotting using antibodies against the CADM1 C-terminus or mouse IgG Fc fragment, respectively. GAPDH was used as a loading control. The graph shows the mean ± standard deviation of relative CADM1 protein levels (CADM1/GAPDH) from three independent experiments. *: p ≤ 0.05 vs. mIgG. (B) Mast cell degranulation was quantified by measuring β-hexosaminidase activity in cell lysates and supernatants. The graph shows the mean (± standard deviation) percentage of β-hexosaminidase release (supernatant/total) from IgE-sensitized BMMCs (top) and CTMCs (bottom) stimulated with phosphate-buffered saline (IgE/PBS) or antigen (IgE/Ag), and from BMMCs (top) and CTMCs (bottom) stimulated with substance P (SP) or left unstimulated (None), across three independent experiments. *: p ≤ 0.05 vs. mIgG.

We further examined whether the presence of fibroblasts influenced CADM1 expression and IgE-mediated CTMC activation. CADM1 protein levels were similarly reduced by cm3E1 treatment in CTMCs cultured either in the presence or absence of NIH/3T3 fibroblasts (Supplementary Figure S2A). In β-hexosaminidase release assays, cm3E1-treated CTMCs exhibited a 20% (p = 0.00045) reduction in IgE-mediated degranulation in the presence of fibroblasts and a 9.5% (p = 0.042) reduction in their absence, relative to mIgG-treated controls (Supplementary Figure S2B).

These results strongly suggest that 3E1 specifically downregulates CADM1 in CTMCs and selectively inhibits IgE-mediated degranulation without affecting non-IgE pathways or BMMCs. This highlights a functional role for CADM1 in regulating IgE-dependent degranulation in dermal-type mast cells.

3E1 treatment altered F-actin rearrangement and reduced secretory granule formation in antigen-activated CTMCs

To further understand how 3E1 modulates IgE-mediated CTMC degranulation, we performed confocal microscopy to visualize intracellular events associated with mast cell activation. Secretory granules were visualized using the styryl dye FM 4–64, and fluorescence intensity was measured at 5, 30, and 60 min following DNP antigen stimulation (Fig. 3). In parallel, we examined the binding of mIgG and cm3E1 to CTMCs using fluorescein-conjugated antibodies, and confirmed that cm3E1 specifically bound to CTMCs, whereas mIgG showed minimal binding (Supplementary Figure S3). In control CTMCs treated with mIgG, FM 4–64 intensity increased in a time-dependent manner after stimulation, indicating active formation of secretory granules. In contrast, CTMCs treated with cm3E1 showed consistently lower FM 4–64 intensity at all time points, most notably at 5 min, where intensity reached approximately 34% of that observed in mIgG-treated cells (p = 0.0066; Fig. 3, Supplementary Figure S4). These results suggest that cm3E1 suppresses secretory granule formation during the early phase of IgE-mediated activation.

Fig. 3.

Effect of 3E1 on secretory granule formation in connective tissue mast cells. Secretory granules were visualized using the styryl dye FM 4–64, and the real-time images were captured at 5, 30, and 60 min after antigen stimulation of connective tissue mast cells (CTMCs). Scale bar: 10 μm. The graph shows the mean ± standard deviation of FM 4–64 fluorescence intensity per cell, quantified from five images per group, across three independent experiments. *: p ≤ 0.05, **: p ≤ 0.01 vs. mouse IgG (mIgG).

Given that the formation, trafficking, and exocytosis of secretory granules are regulated by the cytoskeleton²³, we next examined actin dynamics using rhodamine-conjugated phalloidin staining. F-actin organization was analyzed in CTMCs at 0, 5, 30, and 60 min after DNP antigen stimulation (Fig. 4). In resting cells, F-actin was densely localized at the cell cortex, maintaining a firm cell shape in both mIgG- and cm3E1-treated groups. Upon antigen stimulation, 56%, 79%, and 46% of mIgG-treated CTMCs at 5, 30, and 60 min, respectively, exhibited a reduction in cortical F-actin and a redistribution toward the cytoplasm, forming an internal F-actin meshwork and a ruffled cell shape—a hallmark of cytoskeletal remodeling during degranulation²⁴(Fig. 4). In contrast, this antigen-induced F-actin rearrangement was suppressed in cm3E1-treated CTMCs, with only 13%, 23%, and 26% of cells showing cytoplasmic F-actin redistribution at the corresponding time points (Fig. 4).

Fig. 4.

Effect of 3E1 on F-actin rearrangement in connective tissue mast cells. F-actin organization was assessed in connective tissue mast cells (CTMCs) at rest (0 min) or after antigen stimulation (5, 30, 60 min). Following stimulation, CTMCs showed a reduction in cortical F-actin and redistribution toward the cytoplasm, forming an internal F-actin meshwork and a ruffled cell shape (white arrowheads indicate CTMCs with F-actin rearrangement). Scale bar: 20 μm. The graph shows the mean (± standard deviation) percentage of CTMCs exhibiting F-actin rearrangement, quantified from 10 images per group across duplicate experiments. **: p ≤ 0.01 vs. mIgG.

These findings indicate that cm3E1 inhibits the cytoskeletal reorganization required for granule translocation and degranulation in antigen-stimulated CTMCs.

Intravenous injection of 3E1 resulted in a lower proportion of degranulating mast cells during late-phase response in a mouse model of passive cutaneous anaphylaxis

To examine the effect of the 3E1 antibody on dermal mast cells during acute inflammation, we employed the IgE-mediated passive cutaneous anaphylaxis (PCA) model, a well-established murine model for immediate hypersensitivity reactions. Prior to antigenic challenge on the right ear pinna with 2,4-dinitrofluorobenzene (DNFB), mice were intravenously injected with anti-DNP IgE together with either control chicken–mouse chimeric IgG (cmIgG) or cm3E1.

First, we evaluated the systemic distribution of cm3E1 and cmIgG across major organs and the untreated left pinna using immunoblotting (Supplementary Figure S5). Protein levels were detected with an antibody against the chicken IgY Fab fragment. cm3E1 exhibited relatively high accumulation in the lung and testis, tissues known to express higher levels of CADM1, whereas cmIgG showed no specific tissue distribution (Supplementary Figure S5). Notably, cm3E1 levels in the pinna were 6.54-fold higher than cmIgG, suggesting that cm3E1 preferentially accumulates and is retained in the skin (Supplementary Figure S5).

Four hours after DNFB application, all treated mice (n = 15) exhibited swelling of the right pinna (Table 1), and histological examination of hematoxylin and eosin-stained sections revealed edematous changes (Fig. 5A). To identify dermal mast cells and assess their activation status, tissue sections were stained with toluidine blue. Resting mast cells were characterized by dense, dark staining with tightly packed granules, while degranulating mast cells showed lighter staining with dispersed granules around the cell (Fig. 5A). In sham control mice, the proportion of degranulating mast cells was 28.9%. Following DNFB challenge in cmIgG-treated mice, the proportion was 45.5%. In contrast, cm3E1-treated mice showed a significantly lower proportion of degranulating mast cells (35.1%) compared with the cmIgG group (Fig. 5B; Table 1).

Table 1.

Summary of right ear phenotypes from the passive cutaneous anaphylaxis model ^*: p ≤ 0.05 vs. sham, ^†: p ≤ 0.01 vs. chicken–mouse IgG (cmIgG).

(n)	sham 3	cmIgG 7	cm3E1 8
Ear thickness (mm) Degranulating MCs (%) Tryptase level	0.223 ± 0.00577 28.9 ± 4.01 0.569 ± 0.613	0.407 ± 0.0647* 45.5 ± 6.75* 2.08 ± 0.510*	0.388 ± 0.0287* 35.1 ± 5.24† 1.47 ± 0.583

Fig. 5.

Intravenous 3E1 suppresses dermal mast cell degranulation during late-phase response in a mouse model of passive cutaneous anaphylaxis. (A) Representative histological images of mouse pinnae stained with hematoxylin and eosin (top) or toluidine blue (bottom). Prior to antigenic challenge, mice were intravenously injected with control chicken–mouse chimeric IgG (cmIgG) or chicken–mouse chimeric 3E1 (cm3E1). Sham controls (sham) were injected with anti-dinitrophenol (DNP) IgE but not challenged with antigen. White arrowheads indicate resting mast cells; black arrowheads indicate degranulating mast cells, identified based on morphology. Scale bars: 100 μm (top), 20 μm (bottom). (B) Quantification of degranulating mast cells (%) and tissue tryptase levels in all examined mice. (C) Tissue mast cell tryptase levels measured by immunoblotting. Protein extracts from each group (sham, n = 3; cmIgG, n = 7; and cm3E1, n = 8) were analyzed. β-actin was used as a loading control.

To confirm the histological findings, we measured tissue levels of mast cell tryptase by immunoblotting (Fig. 5C). Tryptase levels were elevated in the pinnae of cmIgG-treated mice compared with those of sham controls. Although tryptase levels in cm3E1-treated mice were lower than those in the cmIgG group, the difference did not reach statistical significance (p = 0.0714) (Fig. 5B; Table 1).

To distinguish immediate-phase mast cell–dependent responses from potential late-phase inflammatory reactions, we additionally evaluated PCA responses at an earlier time point. At 1 h after DNFB application, mice treated with cm3E1 showed a trend toward reduced Evans blue dye extravasation compared with control IgG-treated mice, although this difference did not reach statistical significance (Supplementary Figure S6). Consistently, histological analysis revealed a modest reduction in the percentage of degranulating dermal mast cells in cm3E1-treated mice at this time point, which likewise did not achieve statistical significance (Supplementary Figure S6). These results suggest that cm3E1 may partially attenuate early mast cell activation in vivo, although the magnitude of this effect was limited under the conditions tested.

Discussion

In this study, we demonstrated that CADM1 is upregulated in CTMCs when CTMCs are differentiated from BMMCs, and that 3E1, a monoclonal antibody with high affinity for the extracellular domain of CADM1 ¹⁹, suppresses IgE-mediated mast cell degranulation both in vivo and in vitro. Confocal microscopy further revealed that 3E1 treatment alters F-actin rearrangement and reduces secretory granule formation in antigen-activated CTMCs.

We differentiated CTMCs by coculturing BMMCs with fibroblasts in SCF-supplemented medium. SCF and its receptor KIT are critical regulators of mast cell development and homeostasis²⁵. In addition to providing SCF, fibroblast coculture has been shown to enhance granule maturation, histamine content, and safranin positivity, thereby generating mast cells that more closely resemble tissue-resident CTMCs²⁶. Indeed, coculture also enables more physiologically relevant stromal signals, as recent studies demonstrate that fibroblast–mast cell paracrine interactions fine-tune maturation and effector functions²⁷, highlighting the importance of tissue microenvironmental support for CTMC development in vitro. The microphthalmia transcription factor (MITF), a basic-helix-loop-helix-leucine zipper-type transcription factor, is essential for KIT expression in mast cells²⁸. Previous study has also shown that SCF treatment induces MITF upregulation in BMMCs through KIT signaling²⁹. Compared with other transcription factors such as GATA-binding proteins GATA-1 and GATA-2, MITF is thought to participate in relatively late stage of mast cell differentiation³⁰. In CTMCs, GATA-2 is indispensable for development, whereas downstream MITF is required for histamine synthesis. Because we previously reported that CADM1 expression in mast cell is regulated by MITF⁵, we speculated that SCF-induced CADM1 expression in CTMCs occurs via SCF-KIT-MITF signaling axis.

Importantly, 3E1 selectively downregulated CADM1 protein levels in CTMCs, but not in BMMCs, and inhibited IgE-mediated CTMC degranulation without affecting substance P-induced degranulation. This suggests that suppression of IgE-mediated degranulation is linked to CADM1 downregulation by 3E1. IgE-mediated mast cell activation is triggered by crosslinking of IgE-bound FcεRI with an allergen, leading to FcεRI aggregation within lipid rafts where Lyn kinase phosphorylates FcεRI, initiating downstream signaling³¹. Based on this, we initially hypothesized that 3E1 may suppress FcεRI aggregation at the membrane, but immunoblotting showed no changes in FcεRI protein levels or antigen-induced aggregation in detergent-resistant fractions (Supplementary Figure S7). Thus, mechanisms downstream of FcεRI aggregation are likely responsible for 3E1’s inhibitory effects.

Gaudenzio et al. reported that substance P induces rapid exocytosis of small granules, whereas IgE-dependent activation involves granule–granule fusion and slower release of larger granules³². Cytoskeletal dynamics play a critical role in granule fusion, transport to the plasma membrane, and exocytosis in mast cells²³. In addition, CADM1 interacts with the actin-binding protein 4.1B/DAL-1 via its cytoplasmic domain, and CADM1 downregulation disrupts actin organization and cell morphology in epithelial-like cells^33,34. These observations led us to hypothesize that the inhibitory effect of 3E1 is mediated through CADM1-dependent alterations in the cytoskeleton. Consistent with this hypothesis, confocal microscopy revealed that 3E1 did not affect F-actin distribution in resting CTMCs but significantly impaired antigen-induced F-actin rearrangement. This, coupled with reduced CADM1 levels, suggests that 3E1 suppresses IgE-dependent degranulation by interfering with CADM1-mediated cytoskeletal remodeling during activation. Importantly, our findings align with those of Moiseeva et al., who showed that shRNA-mediated CADM1 knockdown in HMC-1.1 cells increased F-actin filament length even in the absence of activation³⁵. Although the actin patterns in HMC-1.1 cells differed from our CTMC findings—possibly owing to differences in cell type (immortalized versus primary), species, or the use of shRNA versus antibody blockade—both studies support a role for CADM1 in actin regulation in mast cells.

Using spinning disk confocal microscopy, Yokawa et al. demonstrated that CADM1 downregulation in pancreatic α-cells reduces the velocity of secretory granule movement³⁶, implicating CADM1 in granule dynamics, potentially through microtubule-dependent mechanisms. Additionally, Gaudenzio et al. showed that inhibition of IKK-β impairs SNARE complex formation and shifts IgE-driven degranulation toward the rapid, small-granule release pattern typical of substance P stimulation³², implicating IKK-β–mediated SNARE complex formation in granule fusion during IgE-driven degranulation. Using two-photon microscopy, they further demonstrated that IgE-dependent stimulation produces large, ragged granules in dermal mast cells in a PCA mouse model, indicative of granule–granule fusion events³². These findings underscore the importance of cytoskeletal remodeling and granule fusion in IgE-dependent mast cell degranulation. Notably, the absence of an effect of 3E1 on substance P–induced degranulation further supports pathway-specific regulation rather than broad suppression of mast cell activation. Substance P–driven activation proceeds predominantly through rapid exocytosis of small granules with minimal dependence on actin remodeling, whereas IgE-mediated degranulation requires extensive granule–granule fusion and actin-dependent trafficking^23,32. The selective impairment of antigen-induced F-actin rearrangement and granule formation by 3E1 therefore provides a mechanistic explanation for its differential effects on these two activation pathways. Future studies employing high-spatiotemporal-resolution live-cell imaging combined with pharmacologic inhibitors (e.g., IKK-β, actin, or microtubule modulators) and signaling analyses will be crucial to establish a direct mechanistic link among 3E1 treatment, CADM1 downregulation, actin dynamics, and IgE-dependent degranulation.

Traditionally, CADM1 has been studied in the context of mast cell interactions with other cell types, such as neurons and fibroblasts, largely through in vitro coculture³⁷. We previously showed that CADM1 upregulation in IC-2 mast cells enhances neurite adhesion and mast cell activation by neuronal stimuli^9,13, and that CADM1-deficient BMMCs fail to degranulate in response to nerve stimulation¹¹. Similarly, CADM1 knockdown or peptide blockade reduces mast cell adhesion to sensory nerves and impairs IgE-mediated degranulation of BMMCs³⁸. Collectively, these findings underscore CADM1’s role in mast cell–non-mast cell communication as an adhesion molecule. However, our current work demonstrates that CADM1 also regulates IgE-mediated degranulation in monoculture CTMCs, suggesting an intrinsic role for CADM1, independent of cell–cell contact, likely through its influence on cytoskeletal organization. These differences in experimental context are likely critical. CADM1 functions as an adhesion and signaling scaffold whose role is highly dependent on the cellular microenvironment and interacting partners^39,40. In BMMCs cultured alone, where CADM1 expression is relatively low and homophilic or heterophilic ligands are limited, CADM1 may play a minor role in FcεRI-driven degranulation. By contrast, CTMCs differentiated through fibroblast coculture exhibit markedly upregulated CADM1 expression and a connective tissue–type phenotype, rendering them more dependent on CADM1-associated cytoskeletal organization during IgE-mediated activation. Thus, our findings suggest that CADM1 contributes to mast cell degranulation in a stimulus- and context-dependent manner: promoting degranulation in BMMCs during neuron-dependent activation, while modulating IgE-mediated degranulation selectively in CTMCs. Further studies using coculture systems incorporating neurons and fibroblasts, or genetic manipulation of CADM1 in defined mast cell subtypes, will be required to dissect the signaling pathways linking CADM1 expression to F-actin rearrangement and to determine how these pathways integrate adhesion-dependent and adhesion-independent functions.

In a murine model of PCA—a prototypical IgE-dependent type I hypersensitivity reaction in which symptoms are largely mediated by dermal mast cell activation⁴¹—intravenous 3E1 significantly reduced the proportion of degranulating mast cells. To distinguish immediate mast cell–dependent responses from late-phase reactions, we additionally assessed PCA reactions at an early time point using Evans blue dye extravasation. At 1 h after antigen challenge, cm3E1 treatment showed a modest but non-significant tendency toward reduced dye leakage and mast cell degranulation, suggesting a limited impact on the immediate-phase response of PCA under the conditions tested. The more pronounced suppression observed at 4 h, a time point that corresponds to the late-phase reaction of PCA, therefore likely reflects cumulative consequences of impaired mast cell degranulation rather than exclusive inhibition of the immediate hypersensitivity phase.

Unlike 9D2, an antibody that inhibits CADM1 homophilic binding, thereby disrupting mast cell–nerve communication^9,13, 3E1 lacks inhibitory activity against CADM1 binding¹⁸and instead acts directly on mast cell CADM1 without affecting adhesion. This distinction is critical, as PCA reflects acute, mast cell-dominant IgE-driven responses, whereas other mast cell activation pathways—such as substance P–induced activation mediated by Mrgprb2—are highly dependent on neuroimmune interactions. At present, it remains unknown whether 3E1 affects substance P-induced PCA responses in vivo, as such models involve neuronal substance P release and mast cell–nerve signaling, processes in which CADM1 may exert context-dependent functions distinct from IgE-mediated activation. Thus, while 9D2 may be more suitable for conditions in which mast cell–nerve signaling contributes to disease pathogenesis, 3E1’s effects in PCA support its potential as a targeted treatment for acute IgE-mediated dermal hypersensitivity. Future studies should evaluate 3E1 in chronic allergic models, such as those for atopic dermatitis, to determine whether its mast cell-specific actions can provide therapeutic benefit in more complex, multifactorial allergic conditions. Given its high affinity for subcutaneous nerve fibers¹⁹, 3E1 may also hold promise as a locally delivered agent with dual utility in both mast cell inhibition and localized anesthesia.

This study has some limitations. First, we used murine BMMCs and CTMCs; therefore, it remains unclear whether the same mechanisms operate in human mast cells. Further investigations using human mast cells or mast cell lines are needed to confirm the translational relevance of our findings. Second, the effects of 3E1 on reducing CADM1 expression and suppressing degranulation were relatively modest. Additional experiments using genetic approaches such as CADM1 knockdown or overexpression would help clarify the precise role of CADM1 in mast cell degranulation. Third, although 3E1 reduced CADM1 expression and altered cytoskeletal dynamics, the precise molecular mechanism linking CADM1 to FcεRI-mediated signaling remains to be elucidated. Finally, the in vivo analysis was limited to a single acute hypersensitivity model; studies using chronic or systemic allergic models could provide deeper insight into CADM1 function and facilitate the dissection of mast cell heterogeneity across different inflammatory contexts.

In conclusion, our findings demonstrate that CADM1 intrinsically regulates IgE-mediated CTMC activation by supporting cytoskeletal remodeling and secretory granule formation during degranulation. Using the anti-CADM1 monoclonal antibody 3E1, we showed that CADM1 downregulation suppresses IgE-driven mast cell degranulation in vitro and reduces mast cell activation in vivo in a PCA model. Mechanistically, 3E1 impairs F-actin rearrangement and granule dynamics in antigen-activated CTMCs without affecting FcεRI aggregation or non-IgE pathways, highlighting CADM1’s role in mast cell-intrinsic cytoskeletal control.

Methods

Antibodies

The generation of the original chicken monoclonal antibody 3E1 (RRID: AB_592783)⁹, recombinant cm3E1 ¹⁸, and humanized 3E1 ⁴² have been previously described. A control cmIgG was constructed using the variable region of an anti-SARS-CoV-2 S1 IgY antibody (GenBank: UGN74788.1) ⁴³ and the constant region for mouse IgG. Normal mIgG (FUJIFILM Wako Pure Chemical Corp., Osaka, Japan) was used as a control antibody in the cell experiments. Additional antibodies used included those targeting the CADM1 C-terminus (RRID: AB_532287; Sigma-Aldrich, St. Louis, MO, USA), chicken IgY Fab fragment (AB_2339283; Jackson ImmunoResearch, West Grove, PA, USA), mouse IgG Fc fragment (AB_2338454; Jackson ImmunoResearch), human IgG (AB_430835; Promega, Madison, WI, USA), mast cell tryptase (AB_303023; Abcam, Cambridge, UK), Fcε receptor 1α (AB_467707; Thermo Fisher Scientific Inc., Waltham, MA, USA), DNP (DNP-M1, ACRO Biosystems, Newark, DE, USA), c-Kit (AB_2918362; Proteintech, Rosemont, IL, USA), β-actin (AB_10697035; MBL, Aichi, Japan), GAPDH (AB_10699462; MBL), and lamin B (AB_648158; Santa Cruz Biotechnology, Santa Cruz, CA, USA). Secondary antibodies were purchased from Cytiva (Marlborough, MA, USA) and Jackson ImmunoResearch.

Cell culture

NIH/3T3 cells (RRID: CVCL_0594) were cultured in RPMI-1640 medium (FUJIFILM Wako) supplemented with 10% fetal bovine serum. BMMCs were established as described previously¹¹. Briefly, single-cell suspensions from the bone marrow of C57BL/6 mice were cultured for 2–3 weeks in MEMα medium (FUJIFILM Wako) supplemented with 10 ng/ml mouse IL-3 (R&D systems, Minneapolis, MN, USA) and WEHI-3 conditioned medium (Corning, Corning, NY, USA). To induce a CTMC-like phenotype, BMMCs (3.0 × 10⁵ cells per well) were cocultured with mitomycin C-treated NIH/3T3 cells (6.0 × 10⁴ cells per well) in RPMI-1640 medium containing 100 ng/ml mouse stem cell factor (R&D systems), as described previously²⁰with slight modifications. The medium was changed every 2 days, and feeder cells were replaced on day 4. On day 8, cells were harvested by pipetting, centrifuged, resuspended in fresh medium, and reseeded. Cells were transferred to new wells after 2 h to remove remaining feeder cells; the final cell population was considered CTMCs. The purity of mast cells in both BMMC and CTMC cultures was confirmed by flow cytometric analysis of c-Kit and FcεRI expression, as described previously⁴⁴. For toluidine blue staining, cells were fixed in Carnoy’s solution and stained with 0.05% toluidine blue (pH 4.1).

Degranulation assay

To assess IgE-mediated degranulation, BMMCs and CTMCs were sensitized overnight with 0.5 µg/ml anti-DNP IgE. After 6 h of pretreatment with 1 µg/ml cm3E1 or mIgG, cells were washed with Tyrode-HEPES buffer (138 mM NaCl, 2.9 mM KCl, 1 mM CaCl₂, 1 mM MgCl₂, 12 mM NaHCO₃, 0.36 mM NaH₂PO₄, 5 mM glucose, and 10 mM HEPES; pH 7.4) and seeded at 1.0 × 10⁵ cells per well in 96-well plates, either in the presence or absence of mitomycin C-treated NIH/3T3 cells (4.0 × 10³ cells per well). Cells were stimulated with 200 ng/ml DNP-Albumin Conjugate, bovine (DNP-BSA; Sigma-Aldrich) for 1 h. For substance P stimulation, cells were activated with 100 µM substance P (Sigma-Aldrich) for 1 h post-seeding. Degranulation was quantified by measuring β-hexosaminidase activity in both cell lysates and supernatants. Cells were lysed with Tyrode-HEPES buffer containing 0.1% Triton X-100. Samples (50 µl) were incubated with 50 µl of substrate solution (3.3 mM p-nitrophenyl-2-acetoamido-2-deoxy-β-D-glucopyranoside in 0.1 M citrate buffer, pH 4.5) at 37 °C for 70 min. Reactions were terminated with 0.2 M glycine buffer (pH 10.4), and absorbance was measured at 405 nm using a microplate reader (Maltiskan GO, Thermo Fisher).

Confocal imaging

Secretory granules in CTMCs were visualized using the styryl dye FM 4–64 (10 µg/ml; Thermo Fisher) added to DNP-BSA. Fluorescence was captured using a C2 + confocal microscope system (Nikon, Tokyo, Japan), and intensity profiles were analyzed with NIS-Elements software (Nikon). High-magnification images of FM 4-64-labeled CTMCs were additionally obtained using a Fluoview FV3000 confocal microscopy (EVIDENT, Tokyo, Japan). For F-actin staining, CTMCs were fixed with 4% paraformaldehyde at 0, 5, 30, and 60 min after DNP-BSA stimulation, permeabilized with 0.25% Triton X-100, blocked with 2% bovine serum albumin, and stained with phalloidin- rhodamine (Thermo Fisher). Images were acquired using an A1 HD25 confocal microscope system (Nikon).

Passive cutaneous anaphylaxis

The PCA model was established following a previously described protocol⁴⁵. Male BALB/c mice (n = 15, 2 months old, 27–34 g weight; CLEA Japan Inc., Tokyo, Japan) were intravenously injected with 10 µg of anti-DNP mouse monoclonal IgE (AB_259249; Sigma-Aldrich) and 5 mg/kg of either cm3E1 or control cmIgG. After 24 h, 20 µl of 1% (w/v) DNFB (Tokyo Chemical Industry, Tokyo, Japan) dissolved in acetone was applied to the right ear (pinna) under isoflurane anesthesia with a vaporizer (KN-1071-I; Natsume Seisakusho Co., Ltd., Osaka, Japan) at 1 L/min, while the left ear remained untreated. Three littermates served as sham controls and were injected with anti-DNP IgE but received acetone only. Four hours after DNFB application, ear thickness was measured using a digital caliper, and the pinnae were harvested. Each pinna was bisected: one half was fixed with 4% paraformaldehyde phosphate buffer solution (FUJIFILM Wako) for histology, and the other was fresh-frozen for immunoblotting. For assessment of the immediate-phase PCA response, 1% (w/v) Evans blue was intravenously injected simultaneously with DNFB application, and the pinnae were collected 1 h after DNFB challenge. Extravasated Evans blue dye was extracted from the minced pinnae by incubation in formamide at 60 °C overnight and quantified by measuring absorbance at 620 nm using a microplate reader (Tecan, Männedorf, Switzerland). To assess the systemic distribution of cm3E1 and cmIgG, major organs (i.e., brain, heart, lung, liver, spleen, kidney, and testis) and serum were collected from six mice per group. All animal procedures were approved by Kindai University Faculty of Medicine (KAME-2022-084) and performed in accordance with the Animal Care and Use Committee of Kindai University and with ARRIVE guidelines (https://arriveguidelines.org).

Histological examination

Paraffin-embedded pinna sections were stained with hematoxylin and eosin or 0.05% toluidine blue (pH 4.1) using standard methods. Mast cells were identified by their metachromatic staining and classified as resting or degranulated based on morphology. Total and degranulated mast cells were counted across the entire toluidine blue-stained section for all mice (n = 35).

Immunoblotting

Pinnae were minced in 5 volumes of tissue lysis buffer (Cloud-Clone Corp., Houston, TX, USA) and homogenized with a BioMasher (Nippi, Tokyo, Japan). Homogenates were sonicated and centrifuged at 10,000 × g for 5 min, and supernatants were used for analysis. Cell lysate preparation and immunoblotting were performed as previously described¹⁸. Whole un-cropped immunoblot images used in this study were listed in Supplementary Figure S8.

Statistical analysis

Statistical comparisons between two groups were made using an unpaired two-tailed Student’s t-test. For comparisons among three groups, the Steel–Dwass test was applied. A p-value ≤ 0.05 was considered statistically significant.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

M.H.: Data curation, Formal Analysis, Funding acquisition, Investigation, Visualization; A.Y.: Data curation, Formal Analysis, Funding acquisition, Investigation, Writing – Original Draft; F.T. and T.I.: Validation; A.I.: Conceptualization, Funding acquisition, Supervision. All authors reviewed the manuscript.

Funding

This study was supported by the Japan Society for the Promotion of Science Kakenhi (23K06494 to M.H., 22K07034 to A.Y., and 21K06978 to A.I.), Kindai University Research Enhancement Grant (KD2502 to A.Y., KD2401 to A.I.), and the Japan Agency for Medical Research and Development (JP23ym0126809, A.I.).

Data availability

All data generated or analyzed during this study are included in this published article and its supplementary information files. Further details are available from the corresponding authors upon request.

Declarations

Competing interests

A.I. has received research funding from Pharma Foods International, Co., Ltd.